The volume and variety of research in COVID-19 has been steadily growing since throughout the years of the global pandemic. The existing literatures covers a wide range of topics. To gain a thorough understanding of the recent research trend and identify the most valuable highlights from the massive literatures, it is essential to learn and summarize the bibliometric features of existing research and their focus. With the help of deep learning-based text mining techniques, a comprehensive, interdisciplinary insight can be generated from the numerous papers for this crucial and long-lasting crisis. This paper proposes a self-supervised learning method for automatic literature analysis utilizing BERT, an NLP representation technique developed by Google.
Дун Х. A SELF-SUPERVISED LEARNING-BASED AUTOMATED BIBLIOGRAPHY ANALYSIS METHOD // Сборник тезисов докладов конгресса молодых ученых. Электронное издание. – СПб: Университет ИТМО, [2023]. URL: https://kmu.itmo.ru/digests/article/11463